We encountered two cases of AHA in the perioperative period of PD. Case 1 showed a prolonged aPTT preoperatively; however, PD was performed as scheduled. Subsequently, postoperative hemorrhaging continued, and despite hemostatic treatment using rFVIIa and continued immunosuppressive treatment with corticosteroids and rituximab for AHA, the patient eventually died due to subsequent infections. Case 2 developed AHA postoperatively but responded well to corticosteroid treatment. Thus, although these two patients showed similar perioperative conditions during PD, they had markedly different outcomes. Our experience indicates that the recognition and early diagnosis of AHA are important for reducing mortality in the perioperative period of invasive surgery, such as PD.

Patients with AHA present with isolated prolonged aPTT with abnormal bleeding symptoms, such as subcutaneous, muscle, gastrointestinal, genitourinary, or retroperitoneal bleeding [1]. Although more than 50% of cases are idiopathic, AHA possibly develops as a paraneoplastic syndrome [1]. Malignancies were reported to account for 6–18% of the underlying disease in AHA [3, 7,8,9]. Among solid tumors, the incidence of prostate cancer, lung cancer and colon cancer was reported to be high, although there were a few reports of pancreatic cancer and bile duct cancer [10]. Therefore, patients with malignancies could develop AHA both before and after surgery. There are some case reports regarding the development of AHA during the preoperative or postoperative period, including cases involving PD [11,12,13,14,15]. It has also been suggested that AHA may be triggered by a surgical procedure or infections such as cholangitis [12, 15]. Although the mechanism underlying the development of AHA is still unclear, in addition to neoplastic syndrome, an inflammatory state (e.g., from surgery or infection) may trigger the development of AHA. Tumors in the pancreatic head, including pancreatic cancer and bile duct cancer, often lead to cholangitis, cholecystitis, and pancreatitis. Therefore, especially when patients with malignancy develop these inflammatory diseases preoperatively, a systemic examination for bleeding symptoms and reconfirmation of aPTT are desirable. Although postoperative hemorrhage after PD is often caused by POPF, the possibility of AHA should also be considered.

The treatment of AHA reportedly consists of four pillars: prevention of bleeding; treatment of the underlying disease; hemostatic treatment using an FVIII agent, desmopressin, and bypassing agents (rFVIIa or activated prothrombin complex); and immunosuppressive therapy (e.g., steroids, cyclophosphamide and rituximab) for eradication of the inhibitor [1]. To prevent bleeding, invasive procedures should be avoided if AHA develops in the perioperative period. Although treatment of the underlying disease may cure AHA, it has been suggested that surgery should be delayed until inhibitor eradication when feasible [16]. Therefore, when AHA occurs preoperatively, the suitability for surgery should be carefully decided. Regarding hemostatic treatment, the early diagnosis and initiation of treatment are crucial, as a delayed diagnosis was reported to be associated with a poor response or greater consumption of hemostatic agents [17, 18]. Furthermore, although not used in present cases, emicizumab, which is a monoclonal bispecific antibody that mimics the function of FVIII and has been reported to be efficacious for AHA in recent years [19], may be considered for future cases. Immunosuppressive therapy for eradication of the inhibitor may lead to adverse events, such as compromised immunity and delayed wound healing, so careful management is required for postoperative patients. Since the severity of bleeding was reported to be independent of the inhibitor titer and residual FVIII [1], these treatments should be tailored to the clinical symptoms.

In Case 1, the patient supposedly developed AHA at the time of acute cholecystitis before PD, as evidenced by the prolonged aPTT following cholecystitis. Retrospectively, possible causes of the severe clinical course of Case 1 include: PD being carried out as planned without consideration of the prolonged aPTT; the addition of second surgical invasion; the delayed diagnosis and commencement of treatment for AHA (duration from confirmation of the prolonged aPTT to the commencement of AHA treatment: 11 days); and secondary infection during prolonged treatment period due to uncontrollable hemorrhaging. PD should have been postponed, and the cause of prolonged aPTT should have been investigated in detail for Case 1. In Case 2, the patient developed AHA in the early postoperative period after PD. We suspected AHA in this patient due to minor hemorrhaging and prolonged aPTT, based on our experience with Case 1, and initiated early treatment for AHA (duration from confirmation of the prolonged aPTT to the commencement of AHA treatment: 4 days). As a result, hemorrhaging was controlled, and adverse events from steroids were minimal. Thus, the duration from the onset to the initiation of treatment may be crucial. If malignancies and inflammation from surgery or infection can serve as triggers for AHA development, oncological surgeons may encounter AHA more frequently than previously reported. Keeping AHA in mind during the perioperative period is thus important.